{
  "document_id": "lb_qasper_acb71edfecd0645219cfd258956141fe666557f8debad399",
  "family": "public_qasper",
  "split": "test",
  "text": "Introduction and related work\nIn recent years there has been increasing interest on the issue of disinformation spreading on online social media. Global concern over false (or \"fake\") news as a threat to modern democracies has been frequently raised–ever since 2016 US Presidential elections–in correspondence of events of political relevance, where the proliferation of manipulated and low-credibility content attempts to drive and influence people opinions BIBREF0BIBREF1BIBREF2BIBREF3.\nResearchers have highlighted several drivers for the diffusion of such malicious phenomenon, which include human factors (confirmation bias BIBREF4, naive realism BIBREF5), algorithmic biases (filter bubble effect BIBREF0), the presence of deceptive agents on social platforms (bots and trolls BIBREF6) and, lastly, the formation of echo chambers BIBREF7 where people polarize their opinions as they are insulated from contrary perspectives.\nThe problem of automatically detecting online disinformation news has been typically formulated as a binary classification task (i.e. credible vs non-credible articles), and tackled with a variety of different techniques, based on traditional machine learning and/or deep learning, which mainly differ in the dataset and the features they employ to perform the classification. We may distinguish three approaches: those built on content-based features, those based on features extracted from the social context, and those which combine both aspects. A few main challenges hinder the task, namely the impossibility to manually verify all news items, the lack of gold-standard datasets and the adversarial setting in which malicious content is created BIBREF3BIBREF6.\nIn this work we follow the direction pointed out in a few recent contributions on the diffusion of disinformation compared to traditional and objective information. These have shown that false news spread faster and deeper than true news BIBREF8, and that social bots and echo chambers play an important role in the diffusion of malicious content BIBREF6, BIBREF7. Therefore we focus on the analysis of spreading patterns which naturally arise on social platforms as a consequence of multiple interactions between users, due to the increasing trend in online sharing of news BIBREF0.\nA deep learning framework for detection of fake news cascades is provided in BIBREF9, where the authors refer to BIBREF8 in order to collect Twitter cascades pertaining to verified false and true rumors. They employ geometric deep learning, a novel paradigm for graph-based structures, to classify cascades based on four categories of features, such as user profile, user activity, network and spreading, and content. They also observe that a few hours of propagation are sufficient to distinguish false news from true news with high accuracy. Diffusion cascades on Weibo and Twitter are analyzed in BIBREF10, where authors focus on highlighting different topological properties, such as the number of hops from the source or the heterogeneity of the network, to show that fake news shape diffusion networks which are highly different from credible news, even at early stages of propagation.\nIn this work, we consider the results of BIBREF11 as our baseline. The authors use off-the-shelf machine learning classifiers to accurately classify news articles leveraging Twitter diffusion networks. To this aim, they consider a set of basic features which can be qualitatively interpreted w.r.t to the social behavior of users sharing credible vs non-credible information. Their methodology is overall in accordance with BIBREF12, where authors successfully detect Twitter astroturfing content, i.e. political campaigns disguised as spontaneous grassroots, with a machine learning framework based on network features.\nIn this paper, we propose a classification framework based on a multi-layer formulation of Twitter diffusion networks. For each article we disentangle different social interactions on Twitter, namely tweets, retweets, mentions, replies and quotes, to accordingly build a diffusion network composed of multiple layers (on for each type of interaction), and we compute structural features separately for each layer. We pick a set of global network properties from the network science toolbox which can be qualitatively explained in terms of social dimensions and allow us to encode different networks with a tuple of features. These include traditional indicators, e.g. network density, number of strong/weak connected components and diameter, and more elaborated ones such as main K-core number BIBREF13 and structural virality BIBREF14. Our main research question is whether the use of a multi-layer, disentangled network yields a significant advance in terms of classification accuracy over a conventional single-layer diffusion network. Additionally, we are interested in understanding which of the above features, and in which layer, are most effective in the classification task.\nWe perform classification experiments with an off-the-shelf Logistic Regression model on two different datasets of mainstream and disinformation news shared on Twitter respectively in the United States and in Italy during 2019. In the former case we also account for political biases inherent to different news sources, referring to the procedure proposed in BIBREF2 to label different outlets. Overall we show that we are able to classify credible vs non-credible diffusion networks (and consequently news articles) with high accuracy (AUROC up to 94%), even when accounting for the political bias of sources (and training only on left-biased or right-biased articles). We observe that the layer of mentions alone conveys useful information for the classification, denoting a different usage of this functionality when sharing news belonging to the two news domains. We also show that most discriminative features, which are relative to the breadth and depth of largest cascades in different layers, are the same across the two countries.\nThe outline of this paper is the following: we first formulate the problem and describe data collection, network representation and structural properties employed for the classification; then we provide experimental results–classification performances, layer and feature importance analyses and a temporal classification evaluation–and finally we draw conclusions and future directions.\nMethodology ::: Disinformation and mainstream news\nIn this work we formulate our classification problem as follows: given two classes of news articles, respectively $D$ (disinformation) and $M$ (mainstream), a set of news articles $A_i$ and associated class labels $C_i \\in \\lbrace D,M\\rbrace $, and a set of tweets $\\Pi _i=\\lbrace T_i^1, T_i^2, ...\\rbrace $ each of which contains an Uniform Resource Locator (URL) pointing explicitly to article $A_i$, predict the class $C_i$ of each article $A_i$. There is huge debate and controversy on a proper taxonomy of malicious and deceptive information BIBREF1BIBREF2BIBREF15BIBREF16BIBREF17BIBREF3BIBREF11. In this work we prefer the term disinformation to the more specific fake news to refer to a variety of misleading and harmful information. Therefore, we follow a source-based approach, a consolidated strategy also adopted by BIBREF6BIBREF16BIBREF2BIBREF1, in order to obtain relevant data for our analysis. We collected:\nDisinformation articles, published by websites which are well-known for producing low-credibility content, false and misleading news reports as well as extreme propaganda and hoaxes and flagged as such by reputable journalists and fact-checkers;\nMainstream news, referring to traditional news outlets which deliver factual and credible information.\nWe believe that this is currently the most reliable classification approach, but it entails obvious limitations, as disinformation outlets may also publish true stories and likewise misinformation is sometimes reported on mainstream media. Also, given the choice of news sources, we cannot test whether our methodology is able to classify disinformation vs factual but not mainstream news which are published on niche, non-disinformation outlets.\nMethodology ::: US dataset\nWe collected tweets associated to a dozen US mainstream news websites, i.e. most trusted sources described in BIBREF18, with the Streaming API, and we referred to Hoaxy API BIBREF16 for what concerns tweets containing links to 100+ US disinformation outlets. We filtered out articles associated to less than 50 tweets. The resulting dataset contains overall $\\sim $1.7 million tweets for mainstream news, collected in a period of three weeks (February 25th, 2019-March 18th, 2019), which are associated to 6,978 news articles, and $\\sim $1.6 million tweets for disinformation, collected in a period of three months (January 1st, 2019-March 18th, 2019) for sake of balance of the two classes, which hold 5,775 distinct articles. Diffusion censoring effects BIBREF14 were correctly taken into account in both collection procedures. We provide in Figure FIGREF4 the distribution of articles by source and political bias for both news domains.\nAs it is reported that conservatives and liberals exhibit different behaviors on online social platforms BIBREF19BIBREF20BIBREF21, we further assigned a political bias label to different US outlets (and therefore news articles) following the procedure described in BIBREF2. In order to assess the robustness of our method, we performed classification experiments by training only on left-biased (or right-biased) outlets of both disinformation and mainstream domains and testing on the entire set of sources, as well as excluding particular sources that outweigh the others in terms of samples to avoid over-fitting.\nMethodology ::: Italian dataset\nFor what concerns the Italian scenario we first collected tweets with the Streaming API in a 3-week period (April 19th, 2019-May 5th, 2019), filtering those containing URLs pointing to Italian official newspapers websites as described in BIBREF22; these correspond to the list provided by the association for the verification of newspaper circulation in Italy (Accertamenti Diffusione Stampa). We instead referred to the dataset provided by BIBREF23 to obtain a set of tweets, collected continuously since January 2019 using the same Twitter endpoint, which contain URLs to 60+ Italian disinformation websites. In order to get balanced classes (April 5th, 2019-May 5th, 2019), we retained data collected in a longer period w.r.t to mainstream news. In both cases we filtered out articles with less than 50 tweets; overall this dataset contains $\\sim $160k mainstream tweets, corresponding to 227 news articles, and $\\sim $100k disinformation tweets, corresponding to 237 news articles. We provide in Figure FIGREF5 the distribution of articles according to distinct sources for both news domains. As in the US dataset, we took into account censoring effects BIBREF14 by excluding tweets published before (left-censoring) or after two weeks (right-censoring) from the beginning of the collection process.\nThe different volumes of news shared on Twitter in the two countries are due both to the different population size of US and Italy (320 vs 60 millions) but also to the different usage of Twitter platform (and social media in general) for news consumption BIBREF24. Both datasets analyzed in this work are available from the authors on request.\nA crucial aspect in our approach is the capability to fully capturing sharing cascades on Twitter associated to news articles. It has been reported BIBREF25 that the Twitter streaming endpoint filters out tweets matching a given query if they exceed 1% of the global daily volume of shared tweets, which nowadays is approximately $5\\cdot 10^8$; however, as we always collected less than $10^6$ tweets per day, we did not incur in this issue and we thus gathered 100% of tweets matching our query.\nMethodology ::: Building diffusion networks\nWe built Twitter diffusion networks following an approach widely adopted in the literature BIBREF6BIBREF17BIBREF2. We remark that there is an unavoidable limitation in Twitter Streaming API, which does not allow to retrieve true re-tweeting cascades because re-tweets always point to the original source and not to intermediate re-tweeting users BIBREF8BIBREF14; thus we adopt the only viable approach based on Twitter's public availability of data. Besides, by disentangling different interactions with multiple layers we potentially reduce the impact of this limitation on the global network properties compared to the single-layer approach used in our baseline.\nUsing the notation described in BIBREF26. we employ a multi-layer representation for Twitter diffusion networks. Sociologists have indeed recognized decades ago that it is crucial to study social systems by constructing multiple social networks where different types of ties among same individuals are used BIBREF27. Therefore, for each news article we built a multi-layer diffusion network composed of four different layers, one for each type of social interaction on Twitter platform, namely retweet (RT), reply (R), quote (Q) and mention (M), as shown in Figure FIGREF11. These networks are not necessarily node-aligned, i.e. users might be missing in some layers. We do not insert \"dummy\" nodes to represent all users as it would have severe impact on the global network properties (e.g. number of weakly connected components). Alternatively one may look at each multi-layer diffusion network as an ensemble of individual graphs BIBREF26; since global network properties are computed separately for each layer, they are not affected by the presence of any inter-layer edges.\nIn our multi-layer representation, each layer is a directed graph where we add edges and nodes for each tweet of the layer type, e.g. for the RT layer: whenever user $a$ retweets account $b$ we first add nodes $a$ and $b$ if not already present in the RT layer, then we build an edge that goes from $b$ to $a$ if it does not exists or we increment the weight by 1. Similarly for the other layers: for the R layer edges go from user $a$ (who replies) to user $b$, for the Q layer edges go from user $b$ (who is quoted by) to user $a$ and for the M layer edges go from user $a$ (who mentions) to user $b$.\nNote that, by construction, our layers do not include isolated nodes; they correspond to \"pure tweets\", i.e. tweets which have not originated any interactions with other users. However, they are present in our dataset, and their number is exploited for classification, as described below.\nMethodology ::: Global network properties\nWe used a set of global network indicators which allow us to encode each network layer by a tuple of features. Then we simply concatenated tuples as to represent each multi-layer network with a single feature vector. We used the following global network properties:\nNumber of Strongly Connected Components (SCC): a Strongly Connected Component of a directed graph is a maximal (sub)graph where for each pair of vertices $u,v$ there is a path in each direction ($u\\rightarrow v$, $v\\rightarrow u$).\nSize of the Largest Strongly Connected Component (LSCC): the number of nodes in the largest strongly connected component of a given graph.\nNumber of Weakly Connected Components (WCC): a Weakly Connected Component of a directed graph is a maximal (sub)graph where for each pair of vertices $(u, v)$ there is a path $u \\leftrightarrow v$ ignoring edge directions.\nSize of the Largest Weakly Connected Component (LWCC): the number of nodes in the largest weakly connected component of a given graph.\nDiameter of the Largest Weakly Connected Component (DWCC): the largest distance (length of the shortest path) between two nodes in the (undirected version of) largest weakly connected component of a graph.\nAverage Clustering Coefficient (CC): the average of the local clustering coefficients of all nodes in a graph; the local clustering coefficient of a node quantifies how close its neighbours are to being a complete graph (or a clique). It is computed according to BIBREF28.\nMain K-core Number (KC): a K-core BIBREF13 of a graph is a maximal sub-graph that contains nodes of internal degree $k$ or more; the main K-core number is the highest value of $k$ (in directed graphs the total degree is considered).\nDensity (d): the density for directed graphs is $d=\\frac{|E|}{|V||V-1|}$, where $|E|$ is the number of edges and $|N|$ is the number of vertices in the graph; the density equals 0 for a graph without edges and 1 for a complete graph.\nStructural virality of the largest weakly connected component (SV): this measure is defined in BIBREF14 as the average distance between all pairs of nodes in a cascade tree or, equivalently, as the average depth of nodes, averaged over all nodes in turn acting as a root; for $|V| > 1$ vertices, $SV=\\frac{1}{|V||V-1|}\\sum _i\\sum _j d_{ij}$ where $d_{ij}$ denotes the length of the shortest path between nodes $i$ and $j$. This is equivalent to compute the Wiener's index BIBREF29 of the graph and multiply it by a factor $\\frac{1}{|V||V-1|}$. In our case we computed it for the undirected equivalent graph of the largest weakly connected component, setting it to 0 whenever $V=1$.\nWe used networkx Python package BIBREF30 to compute all features. Whenever a layer is empty. we simply set to 0 all its features. In addition to computing the above nine features for each layer, we added two indicators for encoding information about pure tweets, namely the number T of pure tweets (containing URLs to a given news article) and the number U of unique users authoring those tweets. Therefore, a single diffusion network is represented by a vector with $9\\cdot 4+2=38$ entries.\nMethodology ::: Interpretation of network features and layers\nAforementioned network properties can be qualitatively explained in terms of social footprints as follows: SCC correlates with the size of the diffusion network, as the propagation of news occurs in a broadcast manner most of the time, i.e. re-tweets dominate on other interactions, while LSCC allows to distinguish cases where such mono-directionality is somehow broken. WCC equals (approximately) the number of distinct diffusion cascades pertaining to each news article, with exceptions corresponding to those cases where some cascades merge together via Twitter interactions such as mentions, quotes and replies, and accordingly LWCC and DWCC equals the size and the depth of the largest cascade. CC corresponds to the level of connectedness of neighboring users in a given diffusion network whereas KC identifies the set of most influential users in a network and describes the efficiency of information spreading BIBREF17. Finally, d describes the proportions of potential connections between users which are actually activated and SV indicates whether a news item has gained popularity with a single and large broadcast or in a more viral fashion through multiple generations.\nFor what concerns different Twitter actions, users primarily interact with each other using retweets and mentions BIBREF20.\nThe former are the main engagement activity and act as a form of endorsement, allowing users to rebroadcast content generated by other users BIBREF31. Besides, when node B retweets node A we have an implicit confirmation that information from A appeared in B's Twitter feed BIBREF12. Quotes are simply a special case of retweets with comments.\nMentions usually include personal conversations as they allow someone to address a specific user or to refer to an individual in the third person; in the first case they are located at the beginning of a tweet and they are known as replies, otherwise they are put in the body of a tweet BIBREF20. The network of mentions is usually seen as a stronger version of interactions between Twitter users, compared to the traditional graph of follower/following relationships BIBREF32.\nExperiments ::: Setup\nWe performed classification experiments using a basic off-the-shelf classifier, namely Logistic Regression (LR) with L2 penalty; this also allows us to compare results with our baseline. We applied a standardization of the features and we used the default configuration for parameters as described in scikit-learn package BIBREF33. We also tested other classifiers (such as K-Nearest Neighbors, Support Vector Machines and Random Forest) but we omit results as they give comparable performances. We remark that our goal is to show that a very simple machine learning framework, with no parameter tuning and optimization, allows for accurate results with our network-based approach.\nWe used the following evaluation metrics to assess the performances of different classifiers (TP=true positives, FP=false positives, FN=false negatives):\nPrecision = $\\frac{TP}{TP+FP}$, the ability of a classifier not to label as positive a negative sample.\nRecall = $\\frac{TP}{TP+FN}$, the ability of a classifier to retrieve all positive samples.\nF1-score = $2 \\frac{\\mbox{Precision} \\cdot \\mbox{Recall}}{\\mbox{Precision} + \\mbox{Recall}}$, the harmonic average of Precision and Recall.\nArea Under the Receiver Operating Characteristic curve (AUROC); the Receiver Operating Characteristic (ROC) curve BIBREF34, which plots the TP rate versus the FP rate, shows the ability of a classifier to discriminate positive samples from negative ones as its threshold is varied; the AUROC value is in the range $[0, 1]$, with the random baseline classifier holding AUROC$=0.5$ and the ideal perfect classifier AUROC$=1$; thus larger AUROC values (and steeper ROCs) correspond to better classifiers.\nIn particular we computed so-called macro average–simple unweighted mean–of these metrics evaluated considering both labels (disinformation and mainstream). We employed stratified shuffle split cross validation (with 10 folds) to evaluate performances.\nFinally, we partitioned networks according to the total number of unique users involved in the sharing, i.e. the number of nodes in the aggregated network represented with a single-layer representation considering together all layers and also pure tweets. A breakdown of both datasets according to size class (and political biases for the US scenario) is provided in Table 1 and Table 2.\nExperiments ::: Classification performances\nIn Table 3 we first provide classification performances on the US dataset for the LR classifier evaluated on the size class described in Table 1. We can observe that in all instances our methodology performs better than a random classifier (50% AUROC), with AUROC values above 85% in all cases.\nFor what concerns political biases, as the classes of mainstream and disinformation networks are not balanced (e.g., 1,292 mainstream and 4,149 disinformation networks with right bias) we employ a Balanced Random Forest with default parameters (as provided in imblearn Python package BIBREF35). In order to test the robustness of our methodology, we trained only on left-biased networks or right-biased networks and tested on the entire set of sources (relative to the US dataset); we provide a comparison of AUROC values for both biases in Figure 4. We can notice that our multi-layer approach still entails significant results, thus showing that it can accurately distinguish mainstream news from disinformation regardless of the political bias. We further corroborated this result with additional classification experiments, that show similar performances, in which we excluded from the training/test set two specific sources (one at a time and both at the same time) that outweigh the others in terms of data samples–respectively \"breitbart.com\" for right-biased sources and \"politicususa.com\" for left-biased ones.\nWe performed classification experiments on the Italian dataset using the LR classifier and different size classes (we excluded $[1000, +\\infty )$ which is empty); we show results for different evaluation metrics in Table 3. We can see that despite the limited number of samples (one order of magnitude smaller than the US dataset) the performances are overall in accordance with the US scenario. As shown in Table 4, we obtain results which are much better than our baseline in all size classes (see Table 4):\nIn the US dataset our multi-layer methodology performs much better in all size classes except for large networks ($[1000, +\\infty )$ size class), reaching up to 13% improvement on smaller networks ($[0, 100)$ size class);\nIn the IT dataset our multi-layer methodology outperforms the baseline in all size classes, with the maximum performance gain (20%) on medium networks ($[100, 1000)$ size class); the baseline generally reaches bad performances compared to the US scenario.\nOverall, our performances are comparable with those achieved by two state-of-the-art deep learning models for \"fake news\" detection BIBREF9BIBREF36.\nExperiments ::: Layer importance analysis\nIn order to understand the impact of each layer on the performances of classifiers, we performed additional experiments considering separately each layer (we ignored T and U features relative to pure tweets). In Table 5 we show metrics for each layer and all size classes, computed with a 10-fold stratified shuffle split cross validation, evaluated on the US dataset; in Figure 5 we show AUROC values for each layer compared with the general multi-layer approach. We can notice that both Q and M layers alone capture adequately the discrepancies of the two distinct news domains in the United States as they obtain good results with AUROC values in the range 75%-86%; these are comparable with those of the multi-layer approach which, nevertheless, outperforms them across all size classes.\nWe obtained similar performances for the Italian dataset, as the M layer obtains comparable performances w.r.t multi-layer approach with AUROC values in the range 72%-82%. We do not show these results for sake of conciseness.\nExperiments ::: Feature importance analysis\nWe further investigated the importance of each feature by performing a $\\chi ^2$ test, with 10-fold stratified shuffle split cross validation, considering the entire range of network sizes $[0, +\\infty )$. We show the Top-5 most discriminative features for each country in Table 6.\nWe can notice the exact same set of features (with different relative orderings in the Top-3) in both countries; these correspond to two global network propertie–LWCC, which indicates the size of the largest cascade in the layer, and SCC, which correlates with the size of the network–associated to the same set of layers (Quotes, Retweets and Mentions).\nWe further performed a $\\chi ^2$ test to highlight the most discriminative features in the M layer of both countries, which performed equally well in the classification task as previously highlighted; also in this case we focused on the entire range of network sizes $[0, +\\infty )$. Interestingly, we discovered exactly the same set of Top-3 features in both countries, namely LWCC, SCC and DWCC (which indicates the depth of the largest cascade in the layer).\nAn inspection of the distributions of all aforementioned features revealed that disinformation news exhibit on average larger values than mainstream news.\nWe can qualitatively sum up these results as follows:\nSharing patterns in the two news domains exhibit discrepancies which might be country-independent and due to the content that is being shared.\nInteractions in disinformation sharing cascades tends to be broader and deeper than in mainstream news, as widely reported in the literature BIBREF8BIBREF2BIBREF7.\nUsers likely make a different usage of mentions when sharing news belonging to the two domains, consequently shaping different sharing patterns.\nExperiments ::: Temporal analysis\nSimilar to BIBREF9, we carried out additional experiments to answer the following question: how long do we need to observe a news spreading on Twitter in order to accurately classify it as disinformation or mainstream?\nWith this goal, we built several versions of our original dataset of multi-layer networks by considering in turn the following lifetimes: 1 hour, 6 hours, 12 hours, 1 day, 2 days, 3 days and 7 days; for each case, we computed the global network properties of the corresponding network and evaluated the LR classifier with 10-fold cross validation, separately for each lifetime (and considering always the entire set of networks). We show corresponding AUROC values for both US and IT datasets in Figure 6.\nWe can see that in both countries news diffusion networks can be accurately classified after just a few hours of spreading, with AUROC values which are larger than 80% after only 6 hours of diffusion. These results are very promising and suggest that articles pertaining to the two news domains exhibit discrepancies in their sharing patterns that can be timely exploited in order to rapidly detect misleading items from factual information.\nConclusions\nIn this work we tackled the problem of the automatic classification of news articles in two domains, namely mainstream and disinformation news, with a language-independent approach which is based solely on the diffusion of news items on Twitter social platform. We disentangled different types of interactions on Twitter to accordingly build a multi-layer representation of news diffusion networks, and we computed a set of global network properties–separately for each layer–in order to encode each network with a tuple of features. Our goal was to investigate whether a multi-layer representation performs better than one layer BIBREF11, and to understand which of the features, observed at given layers, are most effective in the classification task.\nExperiments with an off-the-shelf classifier such as Logistic Regression on datasets pertaining to two different media landscapes (US and Italy) yield very accurate classification results (AUROC up to 94%), even when accounting for the different political bias of news sources, which are far better than our baseline BIBREF11 with improvements up to 20%. Classification performances using single layers show that the layer of mentions alone entails better performance w.r.t other layers in both countries.\nWe also highlighted the most discriminative features across different layers in both countries; the results suggest that differences between the two news domains might be country-independent but rather due only to the typology of content shared, and that disinformation news shape broader and deeper cascades.\nAdditional experiments involving the temporal evolution of Twitter diffusion networks show that our methodology can accurate classify mainstream and disinformation news after a few hours of propagation on the platform.\nOverall, our results prove that the topological features of multi-layer diffusion networks might be effectively exploited to detect online disinformation. We do not deny the presence of deceptive efforts to orchestrate the regular spread of information on social media via content amplification and manipulation BIBREF37BIBREF38. On the contrary, we postulate that such hidden forces might play to accentuate the discrepancies between the diffusion patterns of disinformation and mainstream news (and thus to make our methodology effective).\nIn the future we aim to further investigate three directions: (1) employ temporal networks to represent news diffusion and apply classification techniques that take into account the sequential aspect of data (e.g. recurrent neural networks); (2) carry out an extensive comparison of the diffusion of disinformation and mainstream news across countries to investigate deeper the presence of differences and similarities in sharing patterns; (3) leverage our network-based features in addition to state-of-the-art text-based approaches for \"fake news\" dete ction in order to deliver a real-world system to detect misleading and harmful information spreading on social media.",
  "char_len": 31978,
  "document_hash": "700d0b98f22b5cef89e45118600cc262bbe5dd9626c05cf5425de82ef03f4839",
  "questions": [
    {
      "question_id": "q00",
      "qtype": "qasper",
      "question": "How is the political bias of different sources included in the model?"
    }
  ]
}